American Indian (Amerind) sign has been used as an alternative or augmentative communication system with members of several clinical populations including adults who have oral verbal apraxia ( 3 ) and nonverbal mentally retarded children (2). While many of the signs in this system are interpretable by untrained observers (particularly those related to basic needs), others have to be taught to persons in a useis environment (home or institution). An approach to providing this training that should require very little time from clinicians is having such persons view a videotape on which the signs are demonstrated, one or more times. This study was intended to assess the effectiveness of such an approach. Ten persons viewed the 193-sign Amerind Video Dictionmy ( 1 ) twice. Four were males and six were females. They ranged in age from 14 to 51 yr. (mean was 25 yr.). None had previous training in manual communication. They represented a variety of occupations including sales, bartending, hairdressing, teaching (preschool), and general office work. The time interval separating the two viewings of the Dictionary was less than one hour for five subjects and of two to four days for the others. Each subject was asked to identify the 193 signs in the Dictionary three times (with the sound turned off so that he or she would not hear the meanings of the signs). They were presented in a different random order each time. Test 1 was administered prior to the first training session to view the Dictionary; Test 2 after the second one; and Test 3 two to three weeks after Test 2. The number of correct identifications on each test was determined for each subject. All 1 0 subjects improved in their ability to understand Amerind signs. The number of signs they identified correctly before viewing the Dictionary ranged from 96 to 115 (mean was 105). Immediately following the second viewing of the Dictionary the number correct ranged from 156 to 186 (mean was 175). The difference between these two sets of scores was significant at the 0.01 level ( 8 = -17.39, d f = 18) . Two to three weeks post-viewing the number of signs identified correctly ranged from 136 to 170 ( M was 157). The difference between this set of scores and the second was significant at the 0.01 level ( t = 3.03, d f = 18) . Though there was some regression, the subjects' abilities to identify signs correctly were still significantly better than they had been on the pretest ( t = 12.67, d f = 18) . These findings suggest that the families of users of Amerind sign and others with whom they communicate can be taught to interpret a high percentage of these signs by viewing a videotape dictionary several times. Since there was some regression, it probably would be desirable to have such persons view the dictionary periodically after the initial training period, particularly if they are not interpreting the signs daily that are demonstrated in it.